Recently, we've covered the topic of continuation betting on our blog, and hopefully, our guide gave you a helpful insight into this important aspect of poker strategy.
Next, we'd like to shed some light on frequent situations (if you are an MTT player) where you might be c-betting too much.
Many players, even experienced ones, tend to gravitate towards c-betting their whole range on many boards. While this may have been a valid strategy a few years ago, it won't be a good approach in 2024.
Don’t get carried away with c-betting low boards
The most common situation in No Limit Hold 'em is the Button versus the Big Blind, which will be our starting point. When you're the preflop raiser, i.e. the Button, the higher the flop structure, the better the board for c-betting; therefore, the Ace high, King high or broadway-heavy boards will be naturally good to put pressure on your opponent.
But what about the lower boards? There are a lot of possible combinations of flops, but thanks to the aggregated flop report option provided by Deepsolver, we can navigate through all of the most representative flop structures.
We did so for 60 flop structures in the BUvsBB scenario on the 40 BB depth, and while on most boards, the IP player can safely c-bet more than 70% of their range, there were a few outliers. Unsurprisingly, all the check-heavy boards included at least two cards lower than nine.
In our case, the most checked-behind boards were 865 with a flush draw, 754 rainbow, A55 rainbow and 752 with a flush draw. Let's look at the first two, 865 FD and 754 R.
For the sake of simplicity, let's force BB to check (which most of your opponents will probably do) and allow the in-position player to use a ⅓ sizing. Under such conditions, the BU should c-bet only with around 50% of their holdings. In such a scenario, the hands you should c-bet include straights, sets, two pairs, many top pairs, overpairs, and combo draws. The hands that are good checking candidates include some potential give up on later streets (like underpairs to the board) but also hands with decent equity, which wouldn't necessarily like to build the pot, like overcards middle or bottom pairs.
In our simulation, the BB should check-raise around 16% of the time, assuming they play game theory optimal. You can toy around with this value to adjust it to what you think represents your average opponent best.
Low cards force the opener to be cautious
Let's look at the second example, the 754 board with the flush draw. In this case, the GTO strategy for BU is to c-bet around 50% of their holdings. There's one "but" though. This board structure represents the small groups of boards where BB should lead about 20% of the time.
Assuming the BB checks 100% (which is very likely, given the population's tendencies), and the BU can use only one sizing, we get a strategy with 63% of checking. What is the reason behind it? The BU is now check-backing more often because the out-of-position player does not lead, meaning their range includes all the nut combos that could lead otherwise.
As a result, the BU player is incentivized to check back more often since their opponent's range is stronger.
Having that in mind, the overall strategy for in-position player is similar to the one on the 865 board; good c-betting candidates are the hands with a lot of showdown value, the best combo draws, and some low equity overcards.
Once again, the checking range consists of give-ups, mediocre pairs and overcards, which hope to realize some of their equity on the showdown.
These two examples represent boards you must be cautious of as the preflop raiser, as you won't get away with c-betting your whole range, at least not versus a competent opponent.
Of course, this is only a tiny fraction of the correct continuation betting strategy, but you can use the same logic and tools we used to determine how to react on other boards in the same situation.
With the aggregated reports option and custom solutions provided by Deepsolver, you can sharpen your continuation betting strategy and adjust it to your opponent's likely actions.
Remember, it's not about memorizing the results of the simulations; it's about understanding what led to such results.
If you're not a Deepsolver user, try it with a seven-day trial. If you've already joined our tribe, visit our Discord server and share your experiences with other members!